ABSTRACT Objective: To identify dietary patterns in an adult population and to verify the factors associated with them. Methods: We conducted a cross-sectional population-based household study with 1,574 individuals aged >20 years, of both sexes, living in Teresina and Picos, Piauí, Brazil. Sociodemographic, lifestyle and food consumption data were collected. Dietary patterns were obtained using principal component analysis. Results: There was a positive and significant association with the healthy eating pattern in women, elderly people, individuals with secondary and higher education and smokers. The white-meat pattern was positively associated with people without a partner and non-white skin, as well as with women, elderly people and smokers, and inversely with individuals with longer screen time and individuals with high school and college education. The unhealthy pattern was positively associated with individuals with excessive screen time and individuals with secondary and higher education, and inversely associated with individuals who were insufficiently active. Conclusion: Women, the elderly, individuals with a partner, non-smokers and active people have a healthier diet. However, men, smokers, with longer screen time and insufficiently active had a more inadequate diet, while the level of education showed contrasting results.